AEBS/LDWS-02-12
091016
DRAFT MINUTES
Preparatory Task Force meeting for the
2nd meeting of GRRF informal group
and
2ndmeeting of GRRF informal group on
Automatic Emergency Braking Systems (AEBS) and
Lane Departure Warning Systems (LDWS)
9-10September 2009 (Preparatory Task Force meeting)
14 September 09 (2nd meetingof GRRF informal group)
Venues:
Preparatory Task Force meeting :German Ministry of Transport, Building and Urban Affairs, Robert-Schuman-Platz1, Bonn, Germany
2nd meeting of GRRF informal group:Palais des Nations, Geneva
Chairman: Mr. Johan Renders (EC) ()
Secretariat:Mr. Olivier Fontaine (OICA)()
Note: These minutes cover the discussions held at both the Preparatory Task Force meeting for the 2nd meeting of GRRF informal group and the 2nd meeting of GRRF informal group on Automatic Emergency Braking Systems (AEBS) and Lane Departure Warning Systems (LDWS). They are an attempt to truly reflect the debates and decisions made at those meetings. In addition, some notes, shall provide the reader with information as to whether a decision was done by the Working Party on Brakes and Running Gears at its 66th session (GRRF-66), which directly followed the two meetings of the informal group mentioned above, concerning some particular questions where guidance was requested.
______
Table of contents
1.-Welcome and Introduction
2.-Approval of the agenda
3.-Review of the Terms of Reference and Rules of Procedure
3.1Existing vs. new regulations
3.2“if fitted” vs. “shall be fitted” approach
3.3Scope
3.4Dates of delivery
4.-Review of the action points from the kick-off meeting in Paris
4.1AEBS
4.2LDWS
5.-Revision of AEBS skeleton paper
5.1Scope and purpose
5.1.1Vehicle categories
5.1.2Collision avoidance vs. collision mitigation
5.2Definitions
5.2.1Advanced Emergency Braking System
5.2.2Other definitions
5.3Application for approval
5.4Specifications
5.4.1General
5.4.2Warning phase
5.4.3Braking phase
5.5Conclusion for AEBS
6.-Revision of the LDWS skeleton paper
6.1Scope and purpose
6.2Definitions
6.3Application for approval
6.4Specifications
6.4.1Test on curved roads
6.4.2Test speed
6.4.3Overriding capabilities
6.4.4Warning indication
6.5Test procedure
6.5.1Test conditions
6.5.2Test course
6.5.3Lane departure warning test
6.5.4Malfunction detection test
6.6Conclusion for LDWS
7.-Date and place of next meetings
AEBS/LDWS-02-12
091016
1.-Welcome and Introduction
The group recognized the challenge of making progress in the discussions during the holiday period between the previous meeting and the present ones. The group also recognized that the Task force meeting is mainly devoted to the technical items, while the regular plenary 2nd meeting will address some political/administrative issues.
See list of attendees as in Annex 1.
2.-Approval of the agenda
Japan withdrew document AEBS/LDWS-02-06 as it is superseded by document AEBS/LDWS-02-10.
The group agreed to include in the future revisions of document AEBS/LDWS-01-14 (list of documents) all documents cited in the reports.
3.-Review of the Terms of Reference and Rules of Procedure
3.1Existing vs. new regulations
Japan favoured the establishment of new regulations instead of amendments to existing regulations, as a help for the Contracting Parties to prepare their national legislations. Germany was concerned that the lead time necessary for editing and establishing new regulations would handicap Industry to make Industrial decisions in time for being ready for the time of application of the systems mandatory equipment. CLEPA supported that point of view, adding that, whatever the decision, it should be done by GRRF well in time for Industry to be ready to prepare design and production of the systems. CLEPA pointed out the particular case of AEBS, where the system is in fact an “add-on” to the existing braking system. The expert from CLEPA added that a lot in the performance of the AEBS depends on the performance of the braking system itself. He however considered the situation less critical for the LDWS. The expert from the European Commission recognized the tight time frame, but clarified that the European institutions have administratively no major time delaysin adoptinga new regulation, compared to an amendment to an existing regulation. He added that, for the sake of clarity, the European Commission’s legal advisersin principle favour the adoption of new regulations in cases similar to the AEBS/LDWS rulemaking. He however committed to internally inquire about the delay differences between the two paths. The United Kingdom was not in favour of inserting the AEBS requirements into the existing braking regulation, UNECE R13, as this already contains two options namely with or without ESC + BAS. The expert from UK thought that GRRF mightbe reluctant to accept a third option to be added with the introduction of AEBS in an additional annex.
Note: GRRF atits 66th session decided that the requirementsfor AEBS as well as for LDWS shall be introduced in the form of new UNECE regulations.
3.2“if fitted” vs. “shall be fitted” approach
Japan was keen to keep the freedom foreach contracting party to mandate the vehicles or classes of vehicles in whichthe advanced safety systems should be fitted. CLEPA asked the contracting parties to consider the manufacturers’ point of view such that Industry could be forced to develop systems for a small number of vehicles, for the one country mandating the system on a particular category only.
The group acknowledged that point of view.
Note: GRRF atits 66th session decided that the provisions for AEBS as well as for LDWS shall be introduced into new UNECE regulations, with the consequence that the systems will be optional in each regulation (“if fitted” approach).
3.3Scope
The Chairreminded that the Terms of Reference (document AEBS/LDWS-01-07) foresee a wide scope in the UNECE regulation, providing the contracting parties the freedom to nationally restrict the scope as necessary in their own territory. In this view, the Chair informed that the European Commission found some merit in the Japanese approach as communicated in document AEBS/LDWS-02-10. Germany, supported by theUnited Kingdom, stressed that the contracting parties themselves are ultimately the final decision makers about the applicability of one regulation in their territory. Some debate took placeconcerning the possible inclusion of the vehicles of categories M2/N2 in the scope of the regulations. While it was well stressed that the informal group will deliver results for those categories in the same time as for the heavier vehicles of catgory M3 and N3, the expert from Daimlercautionedthe group that AEBS currently is nowhere applied on M2/N2, and that defining the requirements for them will be more difficult as there is – unlike for M3 and N3 - no experience with existing systems.
Note: GRRF atits 66th session decided that the scopes for AEBS as well as for LDWS shall be restricted to vehicles of categories 2 and 3, with no derogation in the UNECE regulations.
3.4Dates of delivery
In view of the good progress made on the establishment of a consistent regulatory text on LDWS, the group agreed to advance the target date for delivering to GRRF a draft regulatory text proposal on LDWS, from February 2011 to September 2010.
The target date for delivering to GRRF a draft regulatory text proposal on AEBSremained unchanged to February 2011
4.-Review of the action points from the kick-off meeting in Paris
4.1AEBS
The Chair, as European Commission representative, recalled that provisionsexist in the European General Safety Regulation (GSR) to enable the EC to provide exemptions for certain vehicles and:or classesof vehicles. He invited all European parties to provide their desiderata together with data to justify exemptions for certain vehicles.
Japan presented document AEBS/LDWS-02-09 providing statistical data about accidents involving large trucks in Japan, and some interpretation of the figures.
Daimler committed to provide statistical data at the December 09 meeting (3rd meeting, 3-4 December).
As the author was not present, the presentation of Document AEBS/LDWS-02-05, providing some indication on the proper way to define stationary vehicle vs. moving vehicle, was postponed to the next meeting.
The group also agreed to postpone the debates about failure modes and in-use assessment, as no new input was available.
4.2LDWS
As for the AEBS above, the group agreed to postpone to the next meeting the debates about failure modes and in-use assessment, as no new input was available.
As no input was made available about the definition of the latest warning line; the group decided to defer also this item to the next informal group meeting.
5.-Revision of AEBS skeleton paper
Document:AEBS/LDWS-02-02 (Secretariat)
5.1Scope and purpose
5.1.1Vehicle categories
The group started the discussion by a short debate concerning the format of the skeleton paper (AEBS/LDWS-02-02), as it is formatted as a draft new UNECE regulation rather than a purely structural document, and some delegations challenged this as an anticipation of a decision from the group. The Chairrecalled that the format of the text must simply permit to read the requirements separately, and that there is no intention of reflecting the way the final regulatory document should be presented. He suggested to restrict the discussions to the substance of the document, rather than its format, as this format may change drastically according to the expected guidanceand decisions of GRRF/WP29.
5.1.2Collision avoidance vs. collision mitigation
The Chair then started the debate about how to handle the two different philosophies upon which AEBS currently on the market are based, i.e. movingobstacle and collision avoidance versus stationary obstacle and collision mitigation. He launchedthe idea of basically having a set of requirements for an AEBS system based on the 1st philosophy, then an additional set of requirements for an AEBS system also capable of meeting the other one. Some discussion took place about the way to discriminate the two philosophies. The suppliers were of the opinion that collision avoidance depends too much on the warning means and strategy and on the driver’s reaction, on which the system has no control, and suggested therefore to discuss collision mitigation only. The vehicle manufacturers however challenged that point of view, arguing that there are on the market systems that can avoid the accident without taking into account the driver’s reaction. Japan declared to be ready to consider the development of conditions fora system aiming at collision avoidance. Germany insisted on the need to mandate interaction of the system with the driver. Some suppliers indicated that the main problem is automatic braking in the stationary target scenario. The vehicle manufacturers pointed out that the two philosophies do not differ a lot in technology.
Japan presented document AEBS/LDWS-02-11 (State of the art in Japan) and AEBS/LDWS-02-09 (Statistical data in Japan).
The participants had an exchange of questions and views concerning the Japanese presentations:
- AEBS/LDWS-02-11
- Requirements are different for passenger cars vs. heavy vehicles
- Requirements are the same for stationary vs. moving obstacles
- Uncertainty about accident speed as data rely on police reports and drivers input.
- False warning mainly occur in the area of warning only or low braking phase
- Japanese guidelines have no provisions about brake release.
- Accidents mainly occur on secondary roads, above speed limit, where the speed limiters are of no influence
- General concern about false alarms: legal responsibility if the driver brakes as a reaction to a false alarm.
- Japan in the process of extending guidelines to detection of stationary targets
- Time of braking is defined by the Japanese policy of avoiding any interference of the system with the driver.
- AEBS/LDWS-02-09
- Some experts challenged the conclusion of the presentation stating that the figures justify mandating detection of stationary obstacles.
- Some experts challenged the accuracy of the figures themselves.
- Japan committed to provide the requested clarifications at the 3rd meeting in December about the presentations.
Daimler presented a video showing how theirAEBSsystem works and illustrating the need for a full application of the brakes to avoid collision.
- Obstacle moving at 10km/h, and arriving to a stop, fully laden truck, not braked, colliding at 80km/h i.e. simulating speed reduction of 10km/h.
- Same obstacle, fully laden truck, warning at about 80m distance from the obstacle, start of full braking at about 50/60m distance, deceleration of about 6.25m/s² during about 4s.
- Conditions of test are optimum, no guarantee that the system can perform as well in all real situations
- System does not fit Japanese policy (see above).
- However, interference with driver is unlikely as there are two warnings prior tothe start of automatic braking and driver has constant overriding capabilities.
- 1.6 billion km of positive experience.
Conclusions:
- Confirmation that the AEBS must avoid the collision OR mitigate the effect of the collision
- The group showed some support for the two possibilities for handling the two philosophies:
- Providing provisions for the 2 systems separately (moving target/stationary target)
- Providing one set of requirements covering both systems, as the differences mainly lays on the way to detect the obstacle
- Agreement that the group will
- start byfocusing on the “moving obstacle” scenario;
- Then conduct the same exercise for the “stationary obstacle” scenario;
- Then define the appropriate requirements for decelerations and speeds for both scenarios;
- Then consider the best approach for the text itself.
5.2Definitions
5.2.1Advanced Emergency Braking System
While it was well understood that some guidance would be requested to GRRF, the group debated the wording for thedefinition ofAEBS. The United Kingdomwas keen to have the definition reflecting the event cascade: “Detection/warning/braking”. In this view, the representative of the United Kingdom proposed the following wording:
“Advanced Emergency Braking System means a system which can automatically detect a [potential] forward collision , provide the driver with a warning(s) andactivate the vehicle braking system to decelerate the vehicle with the purpose of avoiding or mitigating a collision.”
Japan suggested to add a reference for avoiding driver over-reliance. Thediscussions also addressed the question whether the definition should contain a reference to the warning, as it is one of the purposes of the system, but withoutmention of performance requirements.
Conclusion: the group did not reach any conclusion about the appropriate wording for the definition of AEBS, but agreed to resume consideration of this item at its next meeting.
5.2.2Other definitions
After having started a systematic revision of the proposed definitions, the group found it more efficient to discuss and/or add the relevant definitions as they might appear necessary during the discussion onthe performance requirements.
5.3Application for approval
Not discussed
5.4Specifications
5.4.1General
The group revised the performance and features of the AEBSsystems currently on the market.
Europe(Daimler)(see document AEBS/LDWS-01-03, slide 5):
- Low level warning via intermittent beep,
- then permanent beep,
- then haptic warning in addition (braking of about 3m/s² to discriminate from Adaptive Cruise Control - ACC). This braking remains as long as until the full braking (above 5m/s²) is necessary
- Full braking occurs very late, such that any driver would have braked before. Daimler has no daily experience of the full braking.
Japan (see document AEBS/LDWS-02-08)
- 1st warning
- 2nd warning
- Braking.
Daimler pointed out that the warning phase is important to keep the driving responsibility in the hands of the driver, and that a haptic warning in the form of a deceleration provides the double advantage of being simultaneously haptic and already evacuating some kinetic energy. While recognizing the merit of a haptic warning provided by a deceleration, some experts challenged this braking as a mandatory feature. Japan informed that the Japanese guidelines only mandate the 1st warning and the braking, and permit the manufacturer to change the 2nd warning into a braking. CLEPA supported this approach.
The group finally agreed on the following principles:
- 2 phases:
- warning
- braking + overriding possibilities
- Warning can be subdivided in a cascade of warnings.
- Subsequent discussions will define whether warning cascade remains at the choice of the manufacturer or must be included in the requirements.
5.4.2Warning phase
The group considered three parameters as being important for defining the requirements for the warning:
- The timing, i.e. when the system should start warningthe driver
- The warning means, whether optical, acoustical or haptic, or a combination of these.
- The overriding capabilities, i.e. what action from the driver should provoke overriding of the system, and what actions should not be permitted as an overriding action.
5.4.2.1The timing
Respecting the Vienna Convention (ChapterII, Article8, Paragraph5) provokes the necessity of including a warning phase to ensure that the driver canalways keep the control of his vehicle, and hence the need for delayingthe intervention of the automatic braking. The European (Daimler)system is based on the principle thatthe warning phase shouldbe at least 2seconds, which means that in certain extreme circumstances the automatic braking phase can be skipped.
The group proposed to base the discussions on the following wording:
“At least [2/1] warning signals (either acoustic, optical or haptic) shall be provided at the latest [TIME] before [a collision is inevitable].”
5.4.2.2The warning means
The European (Daimler)system provides a 1st warning as a combination of acoustic and optical means, then a 2nd warning having three means: acoustic, optical and haptic. The Japanese guidelines request optical and acoustic only.
The group acknowledged that the warning of the driver has two main intentions, i.e. “wake-up”/draw the attention of the driver and let the driver sufficient time to react. Some further discussions should take place to decide whether the last warning of the cascade should be by one or more means. Any earlier warning (by one means) is permitted, not regulated.
5.4.2.1The overriding capabilities
Not discussed
5.4.3Braking phase
Again, the group considered three parameters as being important for defining the requirements for the braking phase:
- The timing, i.e. when the system should start the automatic braking
- The braking power, and the way to assess it (speed reduction, deceleration rate, etc.)
- The overriding capabilities, i.e. what action from the driver should provoke overriding of the system, and what actions should not be permitted as an overriding action
The group could not discuss this item due to lack of time, and agreed to postpone the detailed discussions to the next meeting.